A method for manufacturing a semiconductor memory device includes depositing a bottom metal line layer on a dielectric layer, and patterning the bottom metal line layer into a plurality of bottom metal lines spaced apart from each other. In the method, a plurality of switching element dielectric portions are formed on respective ones of the plurality of bottom metal lines, and a top metal line layer is deposited on the plurality of switching element dielectric portions. The method further includes patterning the top metal line layer into a plurality of top metal lines spaced apart from each other. The plurality of top metal lines are oriented perpendicular to the plurality of bottom metal lines.
Legal claims defining the scope of protection. Each claim is shown in both the original legal language and a plain English translation.
1. A method for manufacturing a semiconductor memory device, comprising: depositing a bottom metal line layer on a first dielectric layer; patterning the bottom metal line layer into a plurality of bottom metal lines spaced apart from each other; forming a plurality of switching element dielectric portions on respective ones of the plurality of bottom metal lines, wherein the switching element dielectric portions are in a stacked configuration with the respective ones of the plurality of bottom metal lines, and lateral sides of the switching element dielectric portions are in-line with lateral sides of the stacked configuration; depositing a second dielectric layer on the first dielectric layer between the plurality of bottom metal lines; forming a barrier metal layer on the second dielectric layer and outer portions of the plurality of switching element dielectric portions thereby exposing a central portion of the plurality of switching element dielectric portions; depositing a top metal line layer on the exposed central portions of the plurality of switching element dielectric portions and over the barrier metal layer; and patterning the top metal line layer into a plurality of top metal lines spaced apart from each other; wherein the plurality of top metal lines are oriented perpendicular to the plurality of bottom metal lines.
This invention relates to semiconductor memory device manufacturing, specifically addressing the challenge of forming a cross-point memory array with precise alignment between perpendicular metal lines and switching elements. The method involves depositing a bottom metal line layer on a first dielectric layer and patterning it into spaced-apart bottom metal lines. Switching element dielectric portions are then formed on each bottom metal line, stacked directly above them with aligned lateral sides. A second dielectric layer is deposited between the bottom metal lines, followed by a barrier metal layer that covers the second dielectric layer and outer regions of the switching element dielectric portions, leaving their central portions exposed. A top metal line layer is deposited over the exposed central portions and the barrier metal layer, then patterned into top metal lines perpendicular to the bottom metal lines. This process ensures self-aligned switching elements between intersecting metal lines, improving manufacturing yield and device performance in cross-point memory structures. The method eliminates the need for complex lithography steps by leveraging the stacked configuration and in-line alignment of the switching elements with the metal lines.
2. The method according to claim 1 , wherein the plurality of top metal lines, the plurality of bottom metal lines, and the plurality of switching element dielectric portions are parts of a memory cell array having a cross-point structure.
The invention relates to a memory cell array with a cross-point architecture, addressing challenges in high-density data storage. The array comprises multiple top metal lines, bottom metal lines, and switching element dielectric portions arranged in a grid. The top and bottom metal lines intersect at multiple points, forming a cross-point structure where memory cells are located. Each memory cell includes a switching element, such as a resistive or phase-change material, encapsulated by dielectric portions to ensure electrical isolation and reliable operation. The cross-point design enables high-density storage by minimizing the footprint of each memory cell, allowing for compact and scalable memory arrays. The dielectric portions around the switching elements prevent electrical interference between adjacent cells, improving data integrity and performance. This structure is particularly useful in non-volatile memory applications, such as resistive RAM (ReRAM) or phase-change RAM (PCRAM), where efficient data storage and retrieval are critical. The invention enhances memory density, reliability, and manufacturing efficiency by optimizing the arrangement and insulation of memory cells in a cross-point architecture.
3. The method according to claim 1 , wherein the switching element dielectric portions are components of respective conductive bridging random access memory devices.
Conductive bridging random access memory (CBRAM) devices are non-volatile memory cells that rely on the formation and disruption of conductive filaments within a dielectric material to store data. A challenge in CBRAM technology is efficiently controlling the switching behavior of these devices, particularly in high-density memory arrays where precise filament formation and reliable switching are critical. The invention addresses this by incorporating dielectric portions within the switching elements of CBRAM devices. These dielectric portions are specifically designed to enhance the switching characteristics, such as reducing variability in switching voltages and improving endurance. The dielectric portions may be composed of materials that facilitate controlled filament growth, such as solid electrolytes or insulating layers with tailored properties. By integrating these dielectric portions into the CBRAM devices, the invention enables more consistent and reliable switching operations, which is essential for high-performance memory applications. The dielectric portions may also be engineered to minimize power consumption and improve scalability, making the technology suitable for advanced memory architectures. This approach ensures that the CBRAM devices maintain stable performance across multiple read/write cycles, addressing key limitations in conventional CBRAM designs.
4. The method according to claim 3 , wherein the switching element dielectric portions comprise at least one of amorphous silicon, amorphous silicon germanium, silicon oxide, hafnium oxide, and aluminum oxide.
This invention relates to semiconductor devices, specifically to switching elements used in memory or logic circuits. The problem addressed is improving the performance and reliability of switching elements by optimizing their dielectric material composition. Traditional switching elements often suffer from leakage currents, high power consumption, or degradation over time due to material limitations. The invention describes a switching element with dielectric portions made from specific materials to enhance electrical insulation and stability. These dielectric portions can be composed of amorphous silicon, amorphous silicon germanium, silicon oxide, hafnium oxide, or aluminum oxide. These materials are chosen for their high dielectric strength, low leakage, and compatibility with semiconductor fabrication processes. The dielectric portions are integrated into the switching element to control current flow, ensuring efficient switching operations while minimizing power loss and device degradation. The use of these materials improves the switching element's ability to maintain a stable insulating state, reducing unintended current leakage and enhancing overall device longevity. This is particularly useful in high-density memory arrays or high-speed logic circuits where reliability and energy efficiency are critical. The invention provides a technical solution to the challenges of material degradation and leakage in semiconductor switching elements, offering a more robust and efficient alternative to conventional designs.
5. The method according to claim 1 , wherein the top metal line layer and the bottom metal line layer each comprise one of copper and silver.
This invention relates to a method for forming a semiconductor device with improved electrical conductivity. The method addresses the challenge of achieving high-performance interconnects in semiconductor devices by using specific conductive materials in key metal line layers. The method involves forming a top metal line layer and a bottom metal line layer, each composed of either copper or silver. These materials are chosen for their excellent electrical conductivity, which enhances signal transmission and reduces resistance in the semiconductor device. The top and bottom metal line layers are part of a larger interconnect structure that includes additional metal line layers and insulating layers. The use of copper or silver in these critical layers improves overall device performance by minimizing signal delay and power loss. The method ensures reliable electrical connections while maintaining compatibility with existing semiconductor fabrication processes. This approach is particularly useful in advanced semiconductor devices where high-speed signal transmission and low resistance are essential for optimal performance.
6. The method according to claim 1 , wherein patterning the bottom metal line layer comprises masking portions of the bottom metal line layer and etching exposed portions of the bottom metal line layer to form the plurality of bottom metal lines.
This invention relates to semiconductor manufacturing, specifically to a method for forming patterned metal lines in integrated circuits. The problem addressed is the precise formation of bottom metal lines in multi-layered interconnect structures, which is critical for ensuring electrical connectivity and performance in advanced semiconductor devices. The method involves patterning a bottom metal line layer by selectively masking portions of the layer and etching the exposed areas to create a plurality of bottom metal lines. The masking step protects specific regions of the metal layer from subsequent etching, while the etching step removes unprotected material to define the desired line patterns. This process ensures accurate and reliable formation of the bottom metal lines, which serve as foundational conductive pathways in the semiconductor device. The invention may also include additional steps such as depositing a dielectric layer over the patterned bottom metal lines and forming vias or interconnects to connect the bottom metal lines to overlying metal layers. The precise patterning of the bottom metal lines is essential for maintaining electrical performance and minimizing signal interference in densely packed semiconductor circuits. The method is particularly useful in advanced semiconductor fabrication where fine-line patterning is required to meet the demands of high-performance integrated circuits.
7. The method according to claim 1 , wherein patterning the top metal line layer comprises masking portions of the top metal line layer and etching exposed portions of the top metal line layer to form the plurality of top metal lines.
This invention relates to semiconductor fabrication, specifically to methods for forming patterned metal lines in integrated circuits. The problem addressed is the precise and efficient formation of top metal lines in multi-layer interconnect structures, which is critical for high-performance and reliable electronic devices. The method involves patterning a top metal line layer to create a plurality of top metal lines. This is achieved by first masking specific portions of the top metal line layer to protect them from subsequent etching. The exposed, unmasked portions of the layer are then etched away, leaving behind the desired patterned metal lines. The masking step ensures that only the intended areas are removed, while the etching step defines the final geometry of the metal lines. The process is part of a broader method for forming a semiconductor device, which includes depositing a top metal line layer over an interlayer dielectric (ILD) and forming a plurality of top metal lines from this layer. The patterned metal lines are then used to connect various components within the semiconductor device, facilitating electrical signal transmission and device functionality. This approach improves manufacturing precision and yield by ensuring accurate patterning of the top metal lines, which is essential for maintaining electrical performance and reliability in advanced semiconductor devices. The method is particularly useful in applications requiring fine-line patterning, such as in high-density interconnect structures.
8. The method according to claim 1 , further comprising forming a plurality of spacers on the plurality of top metal lines.
This invention relates to semiconductor manufacturing, specifically to the formation of spacers on top metal lines in integrated circuits. The problem addressed is the need for precise and reliable spacer formation to enhance device performance and reliability in advanced semiconductor fabrication. The method involves depositing a plurality of top metal lines on a substrate, typically as part of a multi-layer interconnect structure. A conformal spacer material is then deposited over the top metal lines, followed by an anisotropic etching process to remove the spacer material from horizontal surfaces while leaving it on the sidewalls of the top metal lines. This forms a plurality of spacers along the sidewalls of the top metal lines. The spacers can be used to improve electrical isolation, reduce parasitic capacitance, or serve as hard masks for subsequent patterning steps. The spacers are formed from materials such as silicon nitride, silicon oxide, or other dielectric materials, depending on the specific application. The anisotropic etching process ensures that the spacers are precisely defined and uniformly formed along the sidewalls. This technique is particularly useful in advanced semiconductor nodes where feature sizes are extremely small, and precise control of spacer dimensions is critical for device performance. The spacers may also be used to define self-aligned contacts or other features in the semiconductor device.
9. The method according to claim 1 , further comprising: forming a metal cap layer on the bottom metal line layer; forming an electrode layer on the metal cap layer; and patterning the metal cap layer and the electrode layer into a plurality of metal cap layers and a plurality of electrodes corresponding to the plurality of bottom metal lines.
This invention relates to semiconductor device fabrication, specifically to the formation of metal interconnect structures with improved electrical and structural properties. The problem addressed is the need for enhanced conductivity and reliability in metal interconnects, particularly in advanced semiconductor devices where resistance and electromigration issues are critical. The method involves forming a metal cap layer on a bottom metal line layer, which serves as a barrier and adhesion layer to improve the interface between the metal lines and subsequent layers. An electrode layer is then deposited on the metal cap layer, providing electrical contact points. Both the metal cap layer and the electrode layer are patterned into multiple segments corresponding to the underlying bottom metal lines. This patterning ensures precise alignment and electrical isolation between adjacent interconnects, reducing parasitic capacitance and improving signal integrity. The metal cap layer enhances adhesion and prevents diffusion of metal atoms, while the patterned electrode layer enables selective electrical connections. This approach improves device performance by reducing resistance and enhancing reliability in high-density interconnect structures. The method is particularly useful in advanced semiconductor manufacturing where fine-pitch interconnects are required.
10. The method according to claim 9 , wherein the plurality of electrodes are positioned between the plurality of switching element dielectric portions and the plurality of metal cap layers.
A method for fabricating a semiconductor device addresses the challenge of improving electrical performance and reliability in integrated circuits by optimizing the arrangement of conductive and insulating layers. The method involves forming a plurality of switching elements, each with a dielectric portion, and a plurality of metal cap layers. The key innovation is positioning a plurality of electrodes between the switching element dielectric portions and the metal cap layers. This configuration enhances charge carrier mobility and reduces parasitic capacitance, leading to faster switching speeds and lower power consumption. The electrodes are strategically placed to facilitate efficient charge transfer while maintaining structural integrity. The method also includes forming conductive vias to electrically connect the electrodes to other circuit components, ensuring seamless integration into the device architecture. By optimizing the layer stack, the method improves device performance without increasing manufacturing complexity. The technique is particularly useful in advanced semiconductor nodes where minimizing resistance and capacitance is critical for high-speed operation. The resulting device exhibits superior electrical characteristics, making it suitable for applications in high-performance computing, memory devices, and RF circuits.
11. The method according to claim 1 , further comprising depositing an additional dielectric layer between the plurality of top metal lines.
12. A method for manufacturing a semiconductor memory device, comprising: depositing a bottom metal line layer on a first dielectric layer; patterning the bottom metal line layer into a plurality of bottom metal lines spaced apart from each other; forming a plurality of switching element dielectric portions on respective ones of the plurality of bottom metal lines, wherein the switching element dielectric portions are in a stacked configuration with the respective ones of the plurality of bottom metal lines, and lateral sides of the switching element dielectric portions are in-line with lateral sides of the stacked configuration; depositing a second dielectric layer on the first dielectric layer between the plurality of bottom metal lines; forming a barrier metal layer on the second dielectric layer and the plurality of switching element dielectric portions; forming a plurality of openings in the barrier metal layer exposing at least a portion of each of the plurality of switching element dielectric portions; depositing a top metal line layer on the barrier metal layer and in the plurality of openings; and patterning the top metal line layer into a plurality of top metal lines spaced apart from each other.
This invention relates to semiconductor memory device manufacturing, specifically addressing the challenge of integrating switching elements with metal lines in a compact and efficient manner. The method involves depositing a bottom metal line layer on a first dielectric layer and patterning it into spaced-apart bottom metal lines. Switching element dielectric portions are then formed on each bottom metal line, aligned such that their lateral sides are flush with the stacked configuration of the bottom metal lines and dielectric portions. A second dielectric layer is deposited between the bottom metal lines, followed by a barrier metal layer covering both the second dielectric layer and the switching element dielectric portions. Openings are created in the barrier metal layer to expose portions of the switching element dielectric portions. A top metal line layer is deposited over the barrier metal layer and into the openings, then patterned into spaced-apart top metal lines. This approach ensures precise alignment and electrical isolation of the switching elements while maintaining structural integrity and minimizing footprint. The method is particularly useful for high-density memory devices where efficient use of space and reliable electrical connections are critical.
13. The method according to claim 12 , wherein: patterning the bottom metal line layer comprises masking portions of the bottom metal line layer and etching exposed portions of the bottom metal line layer to form the plurality of bottom metal lines; and patterning the top metal line layer comprises masking portions of the top metal line layer and etching exposed portions of the top metal line layer to form the plurality of top metal lines.
This invention relates to semiconductor manufacturing, specifically a method for forming metal interconnect structures in integrated circuits. The problem addressed is the precise and efficient patterning of multiple metal line layers to create reliable electrical connections in advanced semiconductor devices. The method involves forming a bottom metal line layer and a top metal line layer, each separated by an insulating layer. The bottom metal line layer is patterned by masking selected portions and etching the exposed areas to create a plurality of bottom metal lines. Similarly, the top metal line layer is patterned by masking and etching to form a plurality of top metal lines. The patterning process ensures that the metal lines are accurately defined and aligned, which is critical for high-performance semiconductor devices. The insulating layer between the bottom and top metal lines prevents electrical shorts while allowing controlled interconnections where needed. The method ensures precise alignment and spacing of the metal lines, which is essential for minimizing signal interference and improving device reliability. This approach is particularly useful in advanced semiconductor manufacturing where fine-line patterning is required.
14. The method according to claim 12 , wherein the plurality of top metal lines, the plurality of bottom metal lines, and the plurality of switching element dielectric portions are parts of a memory cell array having a cross-point structure.
The invention relates to a memory cell array with a cross-point architecture, addressing challenges in high-density memory design. The array includes multiple top metal lines, multiple bottom metal lines, and multiple switching element dielectric portions. These components form a three-dimensional grid where each intersection between a top metal line and a bottom metal line defines a memory cell. The switching element dielectric portions are positioned between the metal lines to enable selective electrical switching at each cross-point. This structure allows for high-density data storage by minimizing the footprint of each memory cell while maintaining reliable switching performance. The cross-point architecture improves scalability and reduces manufacturing complexity compared to traditional memory designs. The dielectric portions ensure proper insulation and switching functionality, enabling efficient read and write operations. This approach is particularly useful in non-volatile memory applications where compact, high-capacity storage is required. The invention enhances memory density and performance by optimizing the arrangement of conductive lines and dielectric materials in a cross-point configuration.
15. The method according to claim 12 , wherein the switching element dielectric portions are components of respective conductive bridging random access memory devices.
This invention relates to conductive bridging random access memory (CBRAM) devices, which are non-volatile memory technologies that use conductive filaments formed within a dielectric material to store data. A key challenge in CBRAM devices is controlling the formation and dissolution of these conductive filaments to ensure reliable switching between high and low resistance states, which correspond to binary data states. The invention addresses this challenge by incorporating switching element dielectric portions as integral components of CBRAM devices. These dielectric portions are engineered to facilitate precise filament formation and rupture, improving the reliability and endurance of the memory cells. The dielectric portions are designed to interact with conductive electrodes and an electrolyte material, enabling controlled ion migration and filament growth. The invention may also include additional features such as multiple dielectric layers, optimized electrode materials, or specific geometric configurations to enhance performance. By integrating the switching element dielectric portions directly into the CBRAM devices, the invention ensures consistent and predictable switching behavior, reducing variability in memory operations. This approach improves data retention, write/erase cycles, and overall device longevity, making the technology suitable for high-density memory applications. The invention may also include methods for fabricating these integrated structures, ensuring precise control over material properties and device dimensions.
16. The method according to claim 15 , wherein the switching element dielectric portions comprise at least one of amorphous silicon, amorphous silicon germanium, silicon oxide, hafnium oxide, and aluminum oxide.
This invention relates to semiconductor devices, specifically to switching elements used in memory or logic circuits. The problem addressed is improving the performance and reliability of switching elements by optimizing their dielectric material properties. The invention describes a method for fabricating a switching element with dielectric portions made from specific materials to enhance electrical characteristics such as switching speed, leakage current, and endurance. The switching element includes a dielectric layer that controls current flow between two electrodes. The dielectric portions are composed of at least one of amorphous silicon, amorphous silicon germanium, silicon oxide, hafnium oxide, or aluminum oxide. These materials are chosen for their high dielectric strength, low leakage current, and compatibility with semiconductor manufacturing processes. The dielectric portions may be deposited using techniques such as chemical vapor deposition or atomic layer deposition to ensure uniform thickness and precise composition. The invention also involves structuring the dielectric layer to include multiple sub-layers or gradients of these materials to further optimize performance. For example, a combination of silicon oxide and hafnium oxide may be used to balance dielectric strength and switching speed. The switching element may be integrated into a memory cell or logic circuit, where its improved dielectric properties reduce power consumption and increase operational stability. This method ensures reliable switching behavior over extended operational lifetimes.
17. The method according to claim 12 , wherein the top metal line layer and the bottom metal line layer each comprise one of copper and silver.
This invention relates to semiconductor manufacturing, specifically to the formation of metal line layers in integrated circuits. The problem addressed is optimizing electrical conductivity and reliability in multi-layer metal interconnect structures. Traditional interconnects often use aluminum, but copper and silver offer superior conductivity, reducing resistance and improving performance. However, integrating these materials into existing fabrication processes presents challenges, particularly in ensuring compatibility with surrounding dielectric materials and maintaining structural integrity during thermal cycling. The invention describes a method for forming metal line layers in an integrated circuit, where both the top and bottom metal line layers are composed of either copper or silver. These layers are embedded within a dielectric material, forming a stacked interconnect structure. The method ensures proper adhesion and electrical connectivity between the metal lines and adjacent layers, while minimizing defects such as voids or delamination. The use of copper or silver enhances signal transmission speed and reduces power consumption, making the interconnects suitable for high-performance applications. The invention also addresses the challenges of patterning and etching these metals, which can be more difficult than traditional aluminum due to their different chemical properties. The resulting structure provides a robust and efficient interconnect system for advanced semiconductor devices.
18. The method according to claim 12 , further comprising forming a plurality of spacers on the plurality of top metal lines.
This invention relates to semiconductor manufacturing, specifically to the formation of spacers on top metal lines in integrated circuits. The problem addressed is improving the structural integrity and electrical performance of metal interconnects by forming protective spacers around the metal lines. The spacers help prevent short circuits, reduce electromigration, and enhance reliability in advanced semiconductor devices. The method involves depositing a spacer material over a plurality of top metal lines, which are part of a multi-layer interconnect structure. The spacer material is then etched back to form spacers on the sidewalls of the metal lines. The spacers are typically made of dielectric materials such as silicon nitride or silicon oxide, which provide insulation and mechanical support. The process ensures uniform spacer formation across the entire wafer, improving yield and device performance. The spacers are formed after the metal lines are patterned and etched, ensuring precise alignment with the underlying interconnect structure. The method may also include additional steps such as cleaning the metal lines before spacer deposition to improve adhesion and prevent defects. The spacers help maintain critical dimensions and reduce parasitic capacitance, which is crucial for high-speed and high-density semiconductor devices. This technique is particularly useful in advanced nodes where feature sizes are extremely small, and reliability is a major concern.
19. The method according to claim 12 , further comprising: forming a metal cap layer on the bottom metal line layer; forming an electrode layer on the metal cap layer; and patterning the metal cap layer and the electrode layer into a plurality of metal cap layers and a plurality of electrodes corresponding to the plurality of bottom metal lines.
This invention relates to semiconductor device fabrication, specifically to the formation of metal interconnect structures with improved electrical and structural properties. The problem addressed is the need for reliable electrical connections and efficient patterning in advanced semiconductor devices, particularly in memory or logic circuits where precise control of metal layers is critical. The method involves forming a metal cap layer on a bottom metal line layer, which enhances adhesion and conductivity. An electrode layer is then deposited on the metal cap layer. Both layers are patterned into multiple metal cap layers and electrodes, each corresponding to the underlying bottom metal lines. This ensures precise alignment and electrical connectivity between the metal lines and electrodes. The patterning step may involve lithography and etching techniques to define the desired electrode shapes and sizes. The metal cap layer improves the interface between the bottom metal lines and the electrodes, reducing resistance and preventing delamination. The electrodes, once patterned, serve as conductive contacts for subsequent device layers or components. This approach is particularly useful in high-density semiconductor devices where fine feature sizes and reliable interconnects are essential. The method ensures consistent electrical performance and structural integrity across the device.
20. The method according to claim 19 , wherein the plurality of electrodes are positioned between the plurality of switching element dielectric portions and the plurality of metal cap layers.
This invention relates to semiconductor devices, specifically to the arrangement of electrodes in a memory cell structure. The problem addressed is optimizing the electrical and structural integrity of memory cells, particularly in three-dimensional (3D) memory arrays, by improving the positioning and insulation of electrodes relative to switching elements and metal cap layers. The invention describes a method for fabricating a memory cell where a plurality of electrodes are positioned between a plurality of switching element dielectric portions and a plurality of metal cap layers. The switching element dielectric portions electrically insulate the switching elements, which control the flow of current in the memory cell. The metal cap layers provide structural support and electrical connectivity. By placing the electrodes between these two layers, the invention ensures proper electrical isolation while maintaining efficient signal transmission. This arrangement enhances device reliability, reduces leakage current, and improves overall performance in high-density memory arrays. The method involves precise alignment and deposition techniques to ensure the electrodes are correctly positioned without compromising the integrity of the surrounding layers. This solution is particularly useful in advanced memory technologies such as resistive RAM (ReRAM) or phase-change memory (PCM), where precise electrode placement is critical for reliable operation.
Cooperative Patent Classification codes for this invention. Click any code to explore related patents in that topic.
September 20, 2019
March 15, 2022
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.